Semi-supervised regression using cluster ensemble and low-rank co-association matrix decomposition under uncertainties

Vladimir Berikov, Alexander Litvinenko

Результат исследования: Публикации в книгах, отчётах, сборниках, трудах конференцийстатья в сборнике материалов конференциинаучнаярецензирование

Аннотация

In this paper, we solve a semi-supervised regression problem. Due to the luck of knowledge about the data structure and the presence of random noise, the considered data model is uncertain. We propose a method which combines graph Laplacian regularization and cluster ensemble methodologies. The co-association matrix of the ensemble is calculated on both labeled and unlabeled data; this matrix is used as a similarity matrix in the regularization framework to derive the predicted outputs. We use the low-rank decomposition of the co-association matrix to significantly speedup calculations and reduce memory. Numerical experiments using the Monte Carlo approach demonstrate robustness, efficiency, and scalability of the proposed method.

Язык оригиналаанглийский
Название основной публикацииProceedings of the 3rd International Conference on Uncertainty Quantification in Computational Sciences and Engineering, UNCECOMP 2019
РедакторыM. Papadrakakis, V. Papadopoulos, G. Stefanou
ИздательNational Technical University of Athens
Страницы229-242
Число страниц14
ISBN (печатное издание)9786188284494
DOI
СостояниеОпубликовано - 2019
Событие3rd International Conference on Uncertainty Quantification in Computational Sciences and Engineering, UNCECOMP 2019 - Crete, Греция
Продолжительность: 24 июн 201926 июн 2019

Серия публикаций

НазваниеProceedings of the 3rd International Conference on Uncertainty Quantification in Computational Sciences and Engineering, UNCECOMP 2019

Конференция

Конференция3rd International Conference on Uncertainty Quantification in Computational Sciences and Engineering, UNCECOMP 2019
СтранаГреция
ГородCrete
Период24.06.201926.06.2019

Fingerprint Подробные сведения о темах исследования «Semi-supervised regression using cluster ensemble and low-rank co-association matrix decomposition under uncertainties». Вместе они формируют уникальный семантический отпечаток (fingerprint).

  • Цитировать

    Berikov, V., & Litvinenko, A. (2019). Semi-supervised regression using cluster ensemble and low-rank co-association matrix decomposition under uncertainties. В M. Papadrakakis, V. Papadopoulos, & G. Stefanou (Ред.), Proceedings of the 3rd International Conference on Uncertainty Quantification in Computational Sciences and Engineering, UNCECOMP 2019 (стр. 229-242). (Proceedings of the 3rd International Conference on Uncertainty Quantification in Computational Sciences and Engineering, UNCECOMP 2019). National Technical University of Athens. https://doi.org/10.7712/120219.6338.18377